Synergistic Effects of Interleukin-1 , Interleukin-6, and Tumor

Home , Interleukin

Synergistic Effects of Interleukin-1␤, Interleukin-6, and Tumor Necrosis Factor-␣: Central Monoamine, Corticosterone, and Behavioral Variations Karen Brebner, M.S., Shawn Hayley, M.S., Robert Zacharko, Ph.D., Zul Merali, Ph.D., and Hymie Anisman, Ph.D.

The proinflammatory cytokines interleukin-1␤ (IL-1␤), IL-6, these cytokines synergistically increased plasma corticosterone and tumor necrosis factor-alpha (TNF-␣) influence levels. Although IL-1␤ and TNF-␣ provoked variations of neuroendocrine activity, promote central neurotransmitter amine turnover in the hypothalamus, locus coeruleus, and alterations, and induce a constellation of symptoms collectively central amygdala, synergistic effects were not evident in this referred to as sickness behaviors. These cytokines may also elicit respect. Nevertheless, in view of the central amine variations anxiety and anhedonia, and have been associated with induced by the cytokines, it is suggested that immune psychological disturbances in humans. In the present activation may come to influence complex behavioral processes, investigation, systemic IL-1␤ and TNF-␣ dose-dependently as well as affective state. [Neuropsychopharmacology and synergistically disrupted consumption of a highly 22:566–580, 2000] © 2000 American College of palatable food source (chocolate milk), possibly reflecting Neuropsychopharmacology. Published by Elsevier Science Inc. anorexia or anhedonia engendered by the treatments. As well,

KEY WORDS: Cytokines; Corticosterone; Dopamine; alterations may impact on immune activity (Blalock Monoamines; Neurochemical; Norepinephrine; Serotonin; 1994; Dunn 1990; Rivier 1993; Rothwell et al. 1997). It Synergism; IL-1; IL-6; TNF-␣ has been posited that, among other things, the immune It is clear that interactions occur between the immune, system acts like a sensory organ informing the brain of endocrine, central, and autonomic nervous systems. Im- antigenic challenge (Blalock 1994) and that immune ac- munological manipulations (or products of an activated tivation may be interpreted by the CNS as a stressor immune system, e.g., cytokines) may affect neuroendo- (Anisman and Merali 1999; Dunn 1990). Further, cyto- crine and central neurotransmitter processes, and con- kines may be part of a regulatory loop that, by virtue of versely, neuroendocrine and central neurotransmitter effects on CNS functioning, might influence behavioral outputs and may even contribute to the symptoms of behavioral pathologies, including mood and anxiety- From the Institute of Neuroscience, Carleton University, Ottawa, related disorders (Anisman and Merali 1999). Indeed, Ontario, Canada (KB, SH, RZ, HA); and School of Psychology and in humans, depression was associated with variations Department of Cellular and Molecular Medicine (ZM), University of ␤ ␤ Ottawa, Ottawa, Ontario, Canada. of plasma cytokines, including interleukin-1 (IL-1 ), Address correspondence to: Hymie Anisman, Ph.D., Life Sciences IL-1 receptor antagonist (IL-1Ra), IL-2, soluble IL-2 re- Research Building, Carleton University. Ottawa, Ontario K1S 5B6, ceptors, IL-6, and soluble IL-6 receptors (Anisman et al. Canada. Received January 28, 1999; revised May 10, 1999; accepted 1998; Maes 1995; Maes et al. 1995; Muller and Ackenheil December 6, 1999. 1998).

NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 © 2000 American College of Neuropsychopharmacology Published by Elsevier Science Inc. 0893-133X/00/$–see front matter 655 Avenue of the Americas, New York, NY 10010 PII S0893-133X(99)00166-9

NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 567

Cytokines and bacterial endotoxins, such as li- provoke central neurochemical changes, and may thus popolysaccharide (LPS), induce a constellation of ap- also impact on behaviors associated with particular parently adaptive behavioral changes, collectively re- transmitter alterations. Indeed, IL-1␤ was found to act ferred to as “sickness behaviors” (Dantzer et al. 1996). synergistically with a mild stressor to increase For instance, these agents induce fever, reduce social monoamine variations within the prefrontal cortex, nu- exploration, sexual behaviors, and food consumption cleus accumbens and dorsal hippocampus (Merali et al. (Bluthe et al. 1992; Johnson et al. 1996; Plata-Salaman 1997). Moreover IL-1␤ plus TNF-␣ acted synergistically 1988; Plata-Salaman et al. 1988; O’Reilly et al. 1987). In to influence neurotoxicity in mixed neuronal/glial cell addition, endotoxins may induce anxiogenic-like effects cultures containing IFN␥ (Jeohn et al. 1998 ), to stimu- (Lacosta et al. 1999) and disrupt responding for reward- late mitogen-activated protein kinase (Lu et al. 1997), ing brain stimulation (Borowski et al. 1998), possibly re- and along with IL-6 to provoke TNF-␣ mRNA expres- flecting anhedonic effects elicited by the immune chal- sion in rat C6 glioma cells (Gayle et al. 1998). However, lenge. The behavioral effects of endotoxin and unlike these effects, to our knowledge, no such syner- cytokine treatment are paralleled by increased hypo- gisms have been reported with respect to central thalamic-pituitary-adrenal (HPA) activity, as reflected monoamine functioning. The present investigation eval- by increased activity of corticotropin releasing hor- uated the individual and synergistic effects of IL-1␤, IL-6, mone (CRH) and elevated plasma ACTH and corticos- and TNF-␣ on an index of illness behavior (consump- terone levels (Kakucksa et al. 1993; Tilders et al. 1993). tion of a highly preferred food substance, i.e., chocolate Inasmuch as cytokines elicit several effects similar to milk), plasma corticosterone, and central monoamine those of LPS, it has been assumed that at least some of levels and turnover. the endotoxin effects involve IL-1␤, or this cytokine acting conjointly or synergistically with IL-6 and/or TNF-␣ (Dunn 1992a; Ebisui et al. 1994; Long et al. 1990; EXPERIMENTS 1–3 Zanetti et al., 1992; Zhou et al. 1996). Indeed, IL-1␤ and IL-6 synergistically increased plasma corticosterone Individual Effects of IL-1␤, IL-6 and TNF-␣ (Zhou et al. 1996) and ACTH (Matta et al. 1992; Subjects. Male CD-1 mice, eight weeks of age were Perlstein et al. 1991), whereas IL-1␤ and TNF-␣ syner- obtained from Charles River Inc, Laprairie, Quebec, gistically stimulated IL-11 through the production of Canada. After arrival at the facility, mice were housed prostaglandin-E2 in rheumatoid synovial fibroblasts in groups of four and acclimated to the laboratory for at (Mino et al. 1998). As well, IL-1␤ and TNF-␣ synergisti- least two weeks. Room temperature was maintained at cally reduced blood glucose levels (Vogel et al. 1991), 21ЊC, and lighting was maintained on a 12 hr light/ social exploration (Bluthe et al. 1994), and food intake dark cycle (lights on at 0800). One week prior to the be- (Plata-Salaman et al. 1996; Sonti et al. 1996; Yang et al. ginning of the experiments, mice were separated into 1994), and increased the levels of inducible nitric oxide individual polypropylene cages with wire mesh lids synthase (Kuemmerle 1998). and maintained ad libitum on pellet mouse chow (5075 In addition to their peripheral actions, IL-1␤, IL-6, Rodent diet Autoclaved; Ralston Purina) and tap water. and TNF-␣ exert numerous central neurochemical ef- All procedures of the present investigation met the fects. In this respect, IL-1␤ increases NE turnover in the guidelines of the Carleton University Animal Ethics PVN, as well as other hypothalamic nuclei (Dunn Committee, as well as the guidelines set forth by the 1992b; Palazzolo and Quadri, 1992; Shintani et al. 1995). Canadian Council on Animal Care. At all times, efforts Likewise, IL-1␤ increases the accumulation of the sero- were made to minimize the number of animals used, tonin metabolite 5-HIAA in the hypothalamus (Dunn and to minimize animal suffering. 1992b) as well as prefrontal cortex and hippocampus (Zalcman et al. 1994), and may alter hypothalamic DA Procedure. In order to avoid confounding effects re- utilization (Masana et al. 1990; Palazzolo and Quadri lated to the stress of food deprivation, in the present in- 1992; Shintani et al. 1993). Furthermore, IL-1␤ increased vestigation sickness was evaluated by assessing con- in vivo hippocampal serotonergic functioning (Merali et sumption of a highly preferred food substance. Prior to al. 1997; Song et al. 1999), and that of hypothalamic NE, cytokine administration, mice were given free access to 5-HT, and DA release (Shintani et al. 1995). While con- chocolate milk for one hour each day during the light siderably less attention has focused on IL-6 and TNF-␣, phase (1000–1100 hrs). Bottles were weighed at the be- these cytokines have been shown to alter central ginning and end of the session in order to determine monoamine levels, albeit to a lesser degree than that ob- consumption. Following the establishment of a steady served after IL-1␤ treatment (Mefford et al. 1991; Zalc- rate of drinking (three consecutive days over which man et al. 1994). consumption varied by less than 10%), mice received an In addition to their individual effects, it is conceiv- acute i.p. injection of either IL-1␤, IL-6, or TNF-␣ (at one able that IL-1␤, IL-6, and TNF-␣ may synergistically of several doses) or vehicle (pyrogen-free sterile saline)

568 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

in three independent experiments. Consumption was by HPLC using a modification of the method of Seegal measured beginning 45 min (in the case of IL-1␤ and IL-6) et al. (1986). Tissue punches were sonicated in a homog- or 30 min (in the case of TNF-␣) following injection. Re- enizing solution comprised of 14.17g monochloracetic covery data was collected 24 hr following cytokine ad- acid, 0.0186 g disodium EDTA, 5.0 ml methanol, and ministration. 500 ml H20. Using a waters M-6000 pump, guard col- ␤ ␮ Recombinant human IL-1 was kindly provided by umn, radial compression column (5 , C18 reverse Dr. Craig Reynolds (Biological Response Modifiers Pro- phase, 8 mm ϫ 10 cm), and a three cell coulometric elec- gram; National Cancer Institute, Frederick, MD, USA. trochemical detector (ESA Model 5100 A), 20 ␮l of the Produced by E.I. Dupont De Nemours; specific activity ϭ supernatant was passed through the system at a flow 1.81 ϫ 104 U/␮g). Recombinant human TNF-␣ (specific rate of 1.5 ml/min (1400–1600 PSI). The mobile phase activity ϭ 1.1 ϫ 105 U/ ␮g) and IL-6 (specific activity ϭ used for the separation was a modification of that used 7.0 ϫ 103 U/␮g) were obtained from R&D Systems. All by Chiueh et al. (1983). Each liter consisted of 1.3 g of drugs were suspended in pyrogen-free, phosphate heptane sulfonic acid, 0.1 g disodium EDTA, 6.5 ml tri- buffered sterile saline (0.9%). IL-1␤ was administered in ethylamine, and 35 ml acetonitrile. The mobile phase a single dose of either 0.025, 0.05, 0.1, 0.4, or 0.8 ␮g (n ϭ was then filtered (0.22 ␮m filter paper) and degassed 10 per group) in a volume of 0.4 ml. IL-6 was adminis- following which the pH was adjusted to 2.5 with phos- tered in a single dose of 0.1, 0.2, 0.8, or 1.6 ␮g (n ϭ 10 phoric acid. The area and height of the peaks was deter- per group), whereas TNF-␣ was administered in a sin- mined using a Hewlett-Packard integrator. The protein gle dose of 1, 2, or 4 ␮g (n ϭ 10 per group). Vehicle ani- content of each sample was determined using bichin- mals (n ϭ 10 per group) in each experiment were in- choninic acid with a protein analysis kit (Pierce Scien- jected i.p. with 0.4 ml pyrogen-free sterile saline. tific, Brockville, Ontario, Canada) and a spectropho- tometer (Brinkman; PC800 colorimeter). Plasma Corticosterone Assay. In a parallel experiment mice received i.p. administration of either IL-1␤ ␮ ␮ ␣ (0.025, 0.05, or 0.1 g), IL-6 (0.2 or 0.8 g), TNF- (1, 2 Statistical Analysis or 4 ␮g), or their respective vehicles, as described in the behavioral study (n ϭ 10 per group). Mice were decapi- Plasma corticosterone and monoamine and metabolite tated 60 min afterward and trunk blood was collected concentrations for each of the cytokine treatments were in tubes containing 10 ␮l EDTA. The blood was centri- analyzed by one-way analyses of variance, followed by ␣ ϭ fuged at 3500 rpm for 15 min, and the plasma frozen Newman-Keuls multiple comparisons ( 0.05). The and stored at Ϫ80ЊC. The plasma corticosterone concen- chocolate milk consumption on the day prior to cytok- trations were determined, in duplicate, by radioimmu- ine treatment, immediately after, and again 24 h follow- noassay using kits obtained from ICN Biomedicals, Inc. ing cytokine administration, was analyzed by repeated These assays were found to yield less than 10% intra- measures analysis of variance. and interassay variability. Brain Dissection Technique. Brains were rapidly re- EXPERIMENTS 4–6 moved and frozen in isopentane and stored at Ϫ80ЊC until processing. The dissection of the brain was con- Synergistic Effects of IL-1␤, IL-6, and TNF-␣ ducted in a cold chamber, with the brain resting on a As indicated earlier, there is reason to expect that the stage containing dry ice. The brain was placed on a pro-inflammatory cytokines may have synergistic ef- stainless steel dissecting block and razor blades were fects with respect to behavioral and hormonal varia- sequentially slid into adjacent slots (spaced approxi- tions. Using sub-optimal doses of the cytokines based mately 0.5 mm apart) sectioning the brain into a series on the preceding studies, Experiments 4–6 assessed the of coronal sections. Sections were mounted on glass synergistic effect of co-administration of IL-1␤ and IL-6, slides and discrete nuclei were removed by micropunch IL-1␤ and TNF-␣, and IL-6 and TNF-␣ on consumption using a hollow 16 or 20 gauge needle with a bevelled of the palatable chocolate milk solution. Parallel experi- tip, following the mouse brain atlas of Franklin and ments assessed the effects of these treatments on Paxinos (1997). Tissue samples, which included the plasma corticosterone levels, central monoamine levels PVN, locus coeruleus, central amygdala, dorsal hippoc- and turnover. ampus, medial prefrontal cortex, and nucleus accumbens were stored at Ϫ80ЊC for subsequent HPLC deter- mination of the monoamines and their metabolites. Experimental Procedure. A total of 40 experimentally naive, male, CD-1 mice were used in each of the Experi- HPLC Procedure for Analysis of Brain Amine Levels. ments 4–6. The subject characteristics and the housing Levels of DA, NE, and 5-HT and their respective metab- conditions were identical to those described in the pre- olites MHPG, DOPAC, and 5-HIAA were determined ceding studies. In 3 independent studies, mice were as- NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 569

signed to four conditions wherein they received two successive i.p. injections, in a total volume of 0.4 ml. Thus, in each experiment mice received either a single cytokine treatment, two cytokine treatments or only saline treatment. The IL-1␤ was administered in a dose of 0.025 ␮g, IL-6 in a dose of 0.1 ␮g, and TNF-␣ in a dose of 1.0 ␮g. Behavioral testing, as described in the preceding experiments, was conducted 1 hour following drug treatment.

RESULTS Experiments 1–3 Variations of Consumption of Palatable Diet. The administration of both IL-1␤ and TNF-␣ reduced consumption of chocolate milk (see Figures 1a and 1b, respectively). The analysis of variance indicated significant Cytokine ϫ Sampling period interactions for IL-1␤ [F(10,108) ϭ 5.43, p Ͻ .01] and for TNF-␣ [F(6, 72) ϭ 5.97, p Ͻ .01]. The multiple comparisons confirmed that the 0.05, 0.1, 0.2, and 0.8 ␮g doses of IL-1␤ reduced consumption of chocolate milk relative to baseline levels and relative to that of saline treated mice. Likewise, TNF-␣ reduced consumption of the chocolate milk relative to saline-treated mice, but this difference only reached statistical significance at the 4.0 ␮g dose. Con- sumption in all animals tested returned to baseline levels 24 hr following treatment. Among vehicle treated mice consumption did not vary over the sampling periods. In contrast to the effects of IL-1␤ and TNF-␣, systemic administration of IL-6 did not significantly affect consumption (see Figure 1c). Although a small decrease of chocolate milk intake was seen at the 0.1 or 0.2 ␮g doses, this decrease did not approach statistical significance. Indeed, at the higher doses tested (0.8 ␮g and 1.6 ␮g), a modest increase of consumption was detected relative to baseline levels.

Corticosterone Variations. When administered 1 hr prior to testing, IL-1␤ and TNF-␣ dose-dependently influenced plasma corticosterone levels [F(3,36) ϭ Ϯ 10.30, p Ͻ .01 and F(3,76) ϭ 12.88, p Ͻ .01, respec- Figure 1. Mean ( SEM) consumption (ml) of chocolate ␤ tively] (see Table 1). The multiple comparisons indi- milk among mice that received systemic IL-1 (top panel), TNF-␣ (middle panel), or IL-6 (lower panel) (n ϭ 10/group). cated that in mice treated with 0.05 and 0.1 ␮g of IL- ␤ For each cytokine, mice received various doses or saline and 1 , corticosterone levels were significantly increased consumption of milk was measured over a 1 hr period com- relative to saline-treated animals. Treatment with mencing 30 (for TNF-␣) or 45 minutes (for IL-1 and IL-6) ␮ ␤ 0.025 g of IL-1 produced a more modest, but non-sig- afterward. Baseline scores denote average consumption over nificant increase in plasma corticosterone levels rela- the three days immediately preceding cytokine administra- tive to saline-treated mice. Likewise, plasma corti- tion; recovery represents consumption 24 hr after the cytok- costerone levels in mice treated with 4 ␮g of TNF-␣ ine test. were significantly increased when compared to mice treated with saline. The effects of the lower two doses did not differ from one other. In contrast, to the Central Monoamine Variations. One hour after IL-1␤ effects of IL-1␤ and TNF-␣, plasma corticosterone treatment, MHPG levels were influenced in the PVN levels were not affected by the IL-6 treatment. [F(3,35) ϭ 4.92, p Ͻ .05], locus coeruleus [F(3,36) ϭ 4.28, 570 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

Table 1. Plasma Corticosterone Levels (␮g/dl protein) 1 hr after i.p. Cytokine Treatment

Treatment (␮g) Corticosterone

Vehicle 5.84 Ϯ 3.09 IL-1 (0.025) 11.24 Ϯ 1.46 (0.05) 17.75 Ϯ 2.29* (0.1) 21.47 Ϯ 1.30* Vehicle 5.35 Ϯ 1.06 TNF-␣ (1) 6.21 Ϯ 2.35 (2) 6.21 Ϯ 0.97 (4) 17.64 Ϯ 0.97* Vehicle 7.20 Ϯ 2.81 IL-6 (0.1) 6.77 Ϯ 2.03 (0.8) 9.84 Ϯ 1.86

Data expressed as means Ϯ SEM (n ϭ 10); *p Ͻ .05 vs. Vehicle.

p Ͻ .05], and central amygdala [F(3,35) ϭ 2.75, p Ͻ .05]. Post hoc comparisons revealed elevated MHPG levels within the PVN at the 0.05 ␮g dose, in the locus coeruleus in response to 0.025 ␮g dose, and in the central amygdala after the 0.1 ␮g dose (see Figure 2). The levels of NE in these brain regions were not affected by the cytokine treatment. Within the central amygdala NE levels in response to the 0.025 and 0.1 ␮g doses of IL-1␤ (x Ϯ SEM ϭ 21.77 Ϯ 3.80 and 24.12 Ϯ 4.40 ng/mg protein, respectively) were modestly reduced relative to vehicle treated mice (x Ϯ SEM ϭ 31.81 Ϯ 2.54 ng/mg), but this effect was variable and not statistically significant. In addition to the increased NE turnover, IL-1␤ altered the concentrations of 5-HT [F(3,36) ϭ 6.51, p Ͻ .01] and its metabolite, 5-HIAA [F(3,32) ϭ 5.72, p Ͻ .01] within the PVN. The multiple comparisons revealed that the high (0.1 ␮g) IL-1␤ dose increased both 5-HT and 5- HIAA levels within this site (see Figure 3). Additionally, in response to IL-1␤ treatment 5-HIAA accumulation was altered within the central amygdala [F(3,33) ϭ 3.73, p Ͻ .05], and both 5-HT and 5-HIAA accumulation was altered in the prefrontal cortex [F(3,33) ϭ 4.175; 5.40, p Ͻ .01, respectively]. Multiple comparisons revealed that the 0.05 and 0.1 ␮g doses increased the accumulation of me- Figure 2. Mean (Ϯ SEM) MHPG concentrations within the tabolite within the central amygdala. Within the medial PVN, locus coeruleus (LC), and central nucleus of the prefrontal cortex, an elevation of 5-HT was evident in re- amygdala (AMY) 60 minutes following treatment with sev- ␤ ϭ Ͻ sponse to the lowest dose of IL-1␤ (0.025 ␮g), whereas all eral doses of IL-1 (n 8–10/group). * p .05 relative to three doses of IL-1␤ effectively elevated the metabolite saline treated animals. levels (see Figure 3). Levels of DA and the metabolite, DOPAC, were also affected by the IL-1␤ treatment prefrontal cortex, the administration of IL-1␤ affected within the prefrontal cortex [Fs(3,32 and 3,33) ϭ 7.84, neither DA nor DOPAC within the nucleus accumbens. 3.74, p Ͻ .05, respectively]. Post-hoc comparisons deter- Treatment with TNF-␣ elicited region- and neu- mined that, relative to vehicle-treated animals, increased rotransmitter-specific amine variations. In particular, amine and metabolite levels were evident at the 0.025 ␮g levels of NE were not affected by the cytokine in the dose for DA (x Ϯ SEM ϭ 4.64 Ϯ 0.68 vs. 9.7 Ϯ 1.2 ng/mg PVN, central amygdala, or prefrontal cortex, but were protein, respectively) and at the 0.025 and 0.05 ␮g doses significantly reduced in the locus coeruleus [F(3,33) ϭ for DOPAC (x Ϯ SEM ϭ 3.25 Ϯ 0.31 vs. 5.33 Ϯ 0.44 and 4.10, p Ͻ .05] (x Ϯ SEM ϭ 19.86 Ϯ 1.7, 13.23 Ϯ 1.2, 13.76 Ϯ 5.28 Ϯ 0.96 ng/mg protein, respectively). Unlike in the 1.4 and 15.78 Ϯ 1.5 ng/mg protein in the vehicle, 1.0, NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 571

Figure 3. Mean (Ϯ SEM) 5-HT and 5-HIAA concentrations in the PVN, locus coeruleus (LC), and prefrontal cortex (PFC) 60 minutes following systemic IL-1␤ treatment (n ϭ 8–10/group). * p Ͻ .05 vs. saline-treated animals.

2.0, and 4.0 ␮g doses, respectively) and dorsal hippoc- sponse to each of the TNF-␣ doses (x Ϯ SEM ϭ 19.34 Ϯ ampus [F (3,34) ϭ 3.55, p Ͻ .05] (x Ϯ SEM ϭ 6.59 Ϯ 0.45, 1.47, 11.61 Ϯ 2.16, 11.74 Ϯ 1.82 and 11.1 Ϯ 2.42 ng/mg 4.81 Ϯ 0.39, 5.30 Ϯ 0.33 and 5.15 Ϯ 0.45 ng/mg protein, protein for vehicle, 1.0, 2.0, and 4.0 ␮g doses, respec- respectively) following all three doses. In contrast, lev- tively), but increased in the prefrontal cortex (vehicle ϭ els of MHPG were not altered in any of these regions, 2.58 Ϯ 0.28, 1.0 ␮g ϭ 4.17 Ϯ 0.5, 2.0 ␮g ϭ 3.7 Ϯ 0.29 and although a modest elevation (20%) of the metabolite 4 ␮g ϭ 3.4 Ϯ 0.37 ng/mg protein). In neither of these was apparent in the locus coeruleus. brain regions was the altered 5-HT level accompanied Levels of 5-HT were altered by the TNF-␣ treatment by variations of 5-HIAA accumulation. In the hippoc- within the PVN [F(3,26) ϭ 3.52, p Ͻ .05] and prefrontal ampus, neither 5-HT nor 5-HIAA was affected by the cortex [F(3,32) ϭ 2.93, p Ͻ .05]. The multiple compari- cytokine. Finally, TNF-␣ increased nucleus accumbens sons revealed that PVN 5-HT levels were reduced in re- DOPAC accumulation [F(3,29) ϭ 3.71, p Ͻ .05], particu- 572 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

larly at the 2.0 and 4.0 ␮g doses (vehicle ϭ 20.24 Ϯ 1.8, 2.0 ␮g ϭ 32.08 Ϯ 4.3 and 4.0 ␮g ϭ 29.55 Ϯ 2.1 ng/mg protein). Neither DA nor DOPAC accumulation was influenced in any of the other brain regions. Few neurochemical effects were observed in response to systemic IL-6 treatment, although the cytokine was found to alter 5-HT [F(2,16) ϭ 4.61, p Ͻ .05] and DA [F(2,19) ϭ 6.33, p Ͻ .01] concentrations within the central amygdala. Multiple comparisons revealed that while 5-HT levels were raised in a dose-dependent fashion (vehicle ϭ 5.37 Ϯ 0.85, 1.0 ␮g ϭ 7.52 Ϯ 0.46, 0.8 ␮g ϭ 13.07 Ϯ 3.20 ng/mg protein), only the 0.1 ␮g IL-6 dose effectively increased amygdaloid DA levels (vehicle ϭ 16.5 Ϯ 2.1, 0.1 ␮g ϭ 29.65 Ϯ 3.3 ng/mg protein). In contrast to the parent amines, neither 5-HIAA nor DOPAC was affected by the cytokine treatment. Experiments 4–6 Variations in Consumption of a Palatable Diet. Figure 4a shows the chocolate milk consumption as a function of the IL-1␤ and IL-6 treatments. The analysis of variance revealed a significant effect of the IL-1␤ ϫ Sam- pling period interaction [F(2,72) ϭ 8.257, p Ͻ .01]. New- man-Keuls multiple comparisons of the simple effects that comprised the interaction indicated that on Test day there was a modest, but significant, reduction (␣ ϭ 0.05) in the amount of milk consumed by mice treated with 0.025 ␮g IL-1␤ ϩ Vehicle, followed by an increase to baseline levels 24 hrs later. There was no reduction of consumption among vehicle treated animals. Further- more, neither the main effect of IL-6, nor the interactions involving this variable approached significance. There is reason to believe that the synergistic effects of IL-1␤ and IL-6 on HPA activity may be dependent on the timing of cytokine administration. The possibility thus also existed that synergism with respect to illness might be dependent on such a factor. Accordingly, two experiments assessed the conjoint effects of IL-1␤ and IL-6 under conditions wherein animals were tested 30 min after cytokine administration, or where the IL-1␤ treatment preceded that of IL-6. In the first of these studies, the procedure was identical to the preceding Ϯ experiment (n ϭ 8/group) except that testing began 30 Figure 4. Mean ( SEM) consumption of chocolate milk min after cytokine co-administration. In the second ex- among mice that received two systemic injections comprising two cytokines, a single cytokine, or no cytokine treat- periment mice received two injections 30 min apart, ment. The top panel depicts the individual and combined and testing was conducted 1 hr after the second injec- ␮ ϩ ␮ ␤ ␮ effects of 0.025 g IL-1 0.1 g IL-6. The middle panel tion. Thus, mice received IL-1 (0.025 g) followed by shows the individual and combined actions of 0.1 ␮g IL-6 ϩ ␤ ␮ vehicle, or IL-1 followed by IL-6 (0.1 g). 1 ␮g TNF-␣, whereas the lower panel shows the actions of The analyses revealed that neither of these manipu- 0.025 ␮g IL-1 ϩ 1 ␮g TNF-␣ (n ϭ 10/group). Consumption lations culminated in synergy between cytokine treat- was measured over a 1 hr period commencing 1 hr follow- ments. When testing commenced 30 min after their co- ing cytokine co-administration. Baseline scores denote aver- administration, consumption among the groups did age consumption over the three days immediately not differ from one another or from baseline. On Test preceding cytokine administration; recovery represents con- day, IL-1␤ provoked consumption (x Ϯ SEM ϭ 3.07 Ϯ sumption 24 hr after the cytokine test. 0.78 ml) that was moderately reduced relative to baseline (x Ϯ SEM ϭ 3.46 Ϯ 0.31), to vehicle-treated mice (x Ϯ NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 573

SEM ϭ 3.55 Ϯ 0.51), and mice that received IL-6 alone (x Ϯ SEM ϭ 3.25 Ϯ 0.25). The combination of IL-1 ϩ IL- 6 (x Ϯ SEM ϭ 3.18 Ϯ 0.31) did not enhance the effects of IL-1␤ alone. Similarly, when IL-1␤ preceded IL-6 treatment by 30 min, the consumption (x Ϯ SEM ϭ 2.88 Ϯ 0.37) was no different from that of animals that received IL-1␤ in the absence of any other treatment (x Ϯ SEM ϭ 2.96 Ϯ 0.45). In neither case was consumption different from that seen at baseline. The combination of IL-6 (0.1 ␮g) ϩ TNF-␣ (1 ␮g) did not act synergistically to reduce chocolate milk consumption (see Figure 4b). The ANOVA revealed a significant TNF-␣ ϫ Sampling period interaction [F(2,72) ϭ 5.807, p Ͻ .01], and multiple comparisons indicated that on Test day, injection of TNF-␣ reduced chocolate milk consumption relative to vehicle treated mice, but not in comparison to baseline consumption. Treatment with IL-6 did not influence consumption, and likewise the interaction between TNF-␣ and IL-6 was not significant. In contrast to the aforementioned cytokine combinations, consumption was found to vary as a function of the IL-1␤ ϫ TNF-␣ ϫ Sampling period interaction [F(2,72) ϭ 5.175, p Ͻ .01]. Multiple comparisons of the simple effects comprising this interaction confirmed that among vehicle treated animals consumption was stable over sampling periods. The low doses of IL-1␤ and TNF-␣ did not provoke statistically significant changes of consumption. However, the combination of these treatments markedly reduced consumption of chocolate milk relative to baseline scores, as well as in comparison with that of animals that received only one of the cytokines (see Figure 4c). Regardless of the treatment mice received, the consumption exhibited 24 hr afterward was comparable to the baseline scores. In effect, it appears that IL-1␤ ϩ TNF-␣ elicited synergistic effects with respect to consumption of the palatable substance. It will be noted that at low doses the effects of both IL-1␤ and TNF-␣ were variable, as observed with respect to plasma corticosterone concentrations (see corticosterone results), likely reflecting that these doses were just at the threshold to elicit behavioral and neuroendocrine alterations. However, the combination Figure 5. Mean (Ϯ SEM) plasma corticosterone concentra- of the treatments yielded more robust and less variable tions 60 min following co-administration of two cytokines, a effects. single cytokine, or no cytokine. The top panel depicts the individual and combined effects of 0.025 ␮g IL-1 ϩ 0.1 ␮g Corticosterone Variations. Administration of IL-1␤ in- IL-6. The middle panel depicts the individual and combined creased plasma corticosterone concentrations [F(1,36) ϭ actions of 0.1 ␮g IL-6 ϩ 1 ␮g TNF-␣, whereas the actions of 39.82, p Ͻ .01], whereas neither the main effect nor the 0.025 ␮g IL-1 ϩ 1 ␮g TNF-␣ are shown in the lowest panel (n ϭ Ͻ interaction involving IL-6 approached statistical signifi- 10/group). * p .05 relative to vehicle/vehicle treated ani- Њ Ͻ cance (see Figure 5a). Likewise, TNF-␣ increased plasma mals. p .05 relative to mice that received only a single cytokine treatment. corticosterone levels [F(1, 36) ϭ 19.79, p Ͻ .01], and this effect was not enhanced by IL-6 co-administration (see Figure 5b). In contrast, as observed in the behavioral of variance revealed a significant IL-1␤ ϫ TNF-␣ inter- study, the co-administration of IL-1␤ (0.025 ␮g) and action [F(1, 36) ϭ 4.01, p Ͻ .01], and multiple compari- TNF-␣ (1.0 ␮g) synergistically increased plasma corti- sons indicated that the separate administration of either costerone levels (see Figure 5c). Specifically, the analysis IL-1␤ or TNF-␣ was not sufficient to significantly in- 574 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

crease levels of corticosterone relative to vehicle-treated 3.01). However, the treatment effects were variable and animals. However, when administered jointly, the in- did not reach statistical significance. In the case of crease of corticosterone levels was greater than the addi- 5-HIAA, IL-1␤ was again found to provoke a modest but tive effects of the two treatments given alone. non-significant elevation of the metabolite concentration. It should be underscored, however, that in evaluat- Analysis of the combined action of IL-6 and TNF-␣ re- ing the combined effects of IL-1␤ and IL-6, the plasma vealed that 5-HT levels within the PVN declined in re- corticosterone levels elicited by the IL-1␤ treatment sponse to TNF-␣ administration [F(1,27) ϭ 8.89 p Ͻ .01] alone were considerably higher than was observed in (x Ϯ SEM ϭ 11.80 Ϯ 1.04 and 7.98 Ϯ 0.64, for vehicle either the dose response study (Experiment 1) or the and TNF-␣ treated mice, respectively). It was further study evaluating the effects of IL-1␤ ϩ TNF-␣ (Experi- observed that TNF-␣ increased the levels of 5-HIAA, ment 6). The relatively high level of plasma corticoster- but this effect did not reach statistical significance. In one observed in this particular study was surprising, as response to the IL-1␤ and TNF-␣ treatments (Experi- this dose of IL-1␤ ordinarily provoked plasma cortico- ment 6), the analysis revealed that within the PVN, the sterone concentrations in the 10–15 ␮g/dl range (see level of 5-HT varied as a function of the interaction be- Experiment 1, and ensuing experiment). As indicated tween the cytokine treatments [F(1,31) ϭ 4.95, p Ͻ .05]. earlier, the 0.025 ␮g dose of IL-1␤ elicited effects that Multiple comparisons confirmed that IL-1␤ provoked a were just at threshold (see Experiment 1) and thus may reduction of 5-HT, which was attenuated in animals have been subject to greater variability. that received the combination of IL-1␤ and TNF-␣. As indicated earlier, since IL-1␤ and IL-6 may syner- The accumulation of 5-HIAA was not influenced sig- gistically influence HPA activity, depending on the tim- nificantly by IL-1␤ administration, whereas a main ef- ing of their administration (Perlstein et al. 1991; Zhou et fect of the TNF-␣ treatment was observed with respect al. 1996), an additional experiment assessed whether to the metabolite levels [F(1,32) ϭ 6.28, p Ͻ .05] (see Ta- these cytokines would synergistically influence plasma ble 2). A significant interaction between these treat- corticosterone when blood was taken 30 min after treatments was not apparent with respect to 5-HIAA accu- ment. The procedure was identical to that of the preced- mulation. However, it will be noted that while the ing study, except that mice were decapitated 30 min af- elevation of 5-HIAA elicited by TNF-␣ alone did not ter administration of either vehicle (two successive i.p. reach statistical significance, the combination of IL-1␤ ϩ injections), IL-1␤ (0.025 ␮g) plus vehicle, IL-6 (0.1 ␮g) TNF-␣ elicited 5-HIAA levels that exceeded those of the plus vehicle, or IL-1␤ (0.025 ␮g) plus IL-6 (0.1 ␮g). The vehicle treated animals, although not those of mice that analysis of variance indicated that treatment with IL-1␤ received the TNF-␣ alone. increased corticosterone levels [F(1,28) ϭ 8.30 p Ͻ .01]. In Experiment 4 (IL-1␤ ϩ IL-6), the levels of NE and Neither IL-6 nor the interaction between IL-1␤ and IL-6 MHPG within the locus coeruleus did not vary signifi- approached statistical significance (vehicle ϭ 5.57 Ϯ cantly as a function of the cytokine treatments, al- 1.15; IL-1␤ ϭ 13.24 Ϯ 2.57; IL-6 ϭ 7.02 Ϯ 1.90; IL-1␤ ϩ though as shown in Table 3, IL-1␤ alone produced a IL-6 ϭ 13.95 Ϯ 3.75). modest increase of MHPG, coupled with a marginal de- cline of NE concentrations (p Ͻ .1). The analysis of the Central Monoamine Variations. Unlike the synergis- amine and metabolite levels in Experiment 5 (IL-6 ϩ tic effects observed with respect to behavioral and neu- TNF-␣) indicated that treatment with TNF-␣ alone sig- roendocrine variations, within the range of doses used, nificantly increased locus coeruleus MHPG [F(1,33) ϭ the cytokine combinations did not synergistically influ- 13.41, p Ͻ .01], whereas NE levels were unaffected. In ence monoamine levels or turnover, and may even have contrast, IL-6 was uniformly without effect. provoked antagonistic effects. It will be recalled that the doses selected were relatively small and the cytokines generally induced limited effects on amine levels and turnover. Thus, from the outset it must be underscored that the absence of any synergy may be unique to the Table 2. The Effect of IL-1 and TNF-␣ Co-administration on limited range of doses employed in this investigation. the Levels of 5-HT and 5-HIAA in the PVN 1 hr after Treatment In contrast to the monoamine alterations observed Monoamines/Metabolites within the PVN following relatively high cytokine doses, neither NE nor MHPG was affected by the low Treatment 5-HT 5-HIAA ␤ ␣ doses of IL-1 , IL-6, or TNF- administered alone or in Veh ϩ Veh 13.01 Ϯ 1.64 10.39 Ϯ 1.72 combination. In Experiment 4, which assessed the inde- IL-1 (0.025 ␮g) ϩ Veh 8.29 Ϯ 2.01* 11.58 Ϯ 2.43 pendent and conjoint actions of IL-1␤ and IL-6, the level TNF-␣ (1 ␮g) ϩ Veh 11.14 Ϯ 1.25 16.21 Ϯ 2.81 ␮ ϩ ␣ ␮ Ϯ Ϯ of 5-HT was reduced somewhat in animals that re- IL-1 (0.025 g) TNF- (1 g) 15.28 2.75 21.57 4.79* ␤ Ϯ ϭ Ϯ ceived IL-1 treatment alone (x SEM 7.67 1.66) Data expressed as means Ϯ S.E.M. (n ϭ 8–10); *p Ͻ .05 vs. Vehicle. relative to saline treated animals (x Ϯ SEM ϭ 15.33 Ϯ Neurotransmitter concentrations are expressed as ng/mg protein. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 575

Table 3. The Effect of Cytokine Co-administration on the Table 4. Effect of Cytokine Administration on the Levels of Levels of NE and MHPG in the Locus Coeruleus 1 hr Biogenic Amines and Related Metabolites in the Central after Treatment Nucleus of the Amygdala

Monoamines/Metabolites Monoamines/Metabolites

Treatment NE MHPG Treatment NE MHPG

Veh ϩ Veh 22.55 Ϯ 2.84 3.89 Ϯ 0.51 Veh ϩ Veh 19.15 Ϯ 0.77 4.75 Ϯ 0.24 IL-1 (0.025 ␮g) ϩ Veh 13.69 Ϯ 2.09 5.66 Ϯ 0.73 IL-1 (0.025 ␮g) ϩ Veh 19.58 Ϯ 2.30 7.00 Ϯ 0.86* IL-6 (0.1 ␮g) ϩ Veh 24.22 Ϯ 3.86 4.77 Ϯ 0.62 TNF-␣ (1 ␮g) ϩ Veh 21.94 Ϯ 2.72 7.12 Ϯ 0.65* IL-1 (0.025 ␮g) ϩ IL-6 (0.1 ␮g) 24.45 Ϯ 3.17 4.25 Ϯ 0.58 IL-1 (0.025 ␮g) ϩ TNF-␣ (1 ␮g) 19.05 Ϯ 1.26 6.21 Ϯ 0.38 Veh ϩ Veh 21.04 Ϯ 2.60 4.32 Ϯ 0.70 Veh ϩ Veh 23.47 Ϯ 2.98 6.16 Ϯ 0.97 IL-6 (0.1 ␮g) ϩ Veh 21.84 Ϯ 3.08 4.09 Ϯ 0.64 IL-1 (0.025 ␮g) ϩ Veh 26.40 Ϯ 3.29 10.83 Ϯ 2.50* TNF-␣ (1 ␮g) ϩ Veh 17.81 Ϯ 1.83 6.95 Ϯ 0.73* IL-6 (0.1 ␮g) ϩ Veh 22.57 Ϯ 3.40 8.76 Ϯ 0.69 IL-6 (0.1 ␮g) ϩ TNF-␣ (1 ␮g) 21.81 Ϯ 2.09 6.58 Ϯ 0.70* IL-1 (0.025 ␮g)ϩ IL-6 (0.1 ␮g) 22.48 Ϯ 2.87 6.92 Ϯ 0.77 Veh ϩ Veh 19.39 Ϯ 1.40 3.09 Ϯ 0.56 5-HT 5-HIAA IL-1 (0.025 ␮g) ϩ Veh 19.14 Ϯ 2.25 5.18 Ϯ 0.81* Veh ϩ Veh 15.35 Ϯ 1.73 10.55 Ϯ 2.36 TNF-␣ (1 ␮g) ϩ Veh 15.28 Ϯ 2.92 3.82 Ϯ 0.47 IL-1 (0.025 ␮g) ϩ Veh 18.87 Ϯ 3.04 19.51 Ϯ 6.19 IL-1 (0.025 ␮g) ϩ TNF-␣ (1 ␮g) 22.80 Ϯ 2.15 5.46 Ϯ 0.55* IL-6 (0.1 ␮g) ϩ Veh 18.46 Ϯ 1.54 20.05 Ϯ 5.98 IL-1 (0.025 ␮g) ϩ IL-6 (0.1 ␮g) 17.52 Ϯ 2.56 14.72 Ϯ 5.62 Data expressed as means Ϯ S.E.M. (n ϭ 9–10); *p Ͻ .05 vs. Vehicle (Veh). Neurotransmitter concentrations are expressed as ng/mg protein. Data expressed as means Ϯ S.E.M. (n ϭ 8–10); *p Ͻ .05 vs. Vehicle (Veh). Neurotransmitter concentrations are expressed as ng/mg protein.

Finally, in Experiment 6 (IL-1␤ ϩ TNF-␣), the 0.025 ␮g dose of IL-1␤ significantly elevated MHPG levels in ment where IL-6 was administered in combination with the locus coeruleus [F(1,34) ϭ 9.54, p Ͻ .01], whereas TNF-␣, the IL-6 alone was found to increase levels of the effect of TNF-␣ was not statistically significant. 5-HIAA [F(1,25) ϭ 4.557 p Ͻ .05] (vehicle ϭ 9.27 Ϯ 1.73, Mice that received both IL-1␤ ϩ TNF-␣ exhibited and 0.1 ␮g ϭ 13.71 Ϯ 2.34 ng/mg protein), but this ef- MHPG levels that exceeded that of vehicle treated ani- fect was not modified by the addition of TNF-␣ (16.48 Ϯ mals; however, MHPG was not synergistically affected 4.08 ng/mg protein). Furthermore TNF-␣ did not inter- in response to the combination of the cytokine treat- act with IL-6 to alter monoamine levels or turnover in ments (see Table 3). any of the other brain regions investigated. Finally, in as- The effect of the cytokines on NE activity within the sessing the effects of IL-1␤ ϩ TNF-␣, it was observed that central amygdala was relatively marked. While the co- although IL-1␤ elicited a modest elevation of 5-HIAA ac- administration of IL-1␤ ϩ IL-6 did not influence NE cumulation, this effect did not approach statistical sig- levels within the central amygdala, a significant interac- nificance, and was not further enhanced by co-adminis- tion between these cytokines was observed with respect tration of TNF-␣. to MHPG accumulation [F(1,34) ϭ 4.65, p Ͻ .05]. Analy- sis of the simple effects comprising the interaction showed that IL-1␤ increased levels of the metabolite, DISCUSSION whereas IL-6 was without effect. Curiously, the effects Behavioral and Neuroendocrine Variations of the IL-1␤ treatment were attenuated in mice that had also received IL-6 administration. Analysis of the con- It will be recalled that cytokines may engender a sick- joint action of IL-1␤ and TNF-␣ indicated that within ness profile (Kent et al. 1996), as well as neuroendocrine this region MHPG levels varied as a function of the IL-1␤ ϫ and central neurotransmitter changes reminiscent of TNF-␣ interaction [F(1,33) ϭ 7.53, p Ͻ .05]. Analysis of those elicited by stressors (Anisman and Merali 1999; the simple effects comprising the interaction revealed Dunn 1988, 1990). In fact, the possibility has been con- that both IL-1␤ and TNF-␣ administered individually sidered that immunological challenges and the pres- increased MHPG levels. However, a further increase of ence of elevated circulating cytokine levels might be in- MHPG was not elicited by the cytokine co-administra- terpreted as a stressor (Anisman and Merali 1999; Dunn tion (see Table 4). 1988). In accordance with the reports that cytokines The 5-HIAA accumulation within the central may elicit sickness (Kent et al. 1996), systemically ad- amygdala in the experiments assessing the synergistic ministered IL-1␤ and TNF-␣ suppressed the consump- actions of the cytokines was found to be unusually vari- tion of a highly palatable substance, possibly reflecting able, and hence must be interpreted cautiously. Al- either an anorexic or an anhedonic response to cytokine though IL-1␤ and IL-6 both produced almost two-fold administration. In contrast, IL-6 appeared to be without increases of 5-HIAA accumulation, these effects were effect in this respect. The effects of IL-1␤ and TNF-␣ nonetheless nonsignificant (see Table 4). In the experi- tended to be pronounced at the higher doses, but were 576 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

relatively variable at the lower doses. Across experi- was collected 30 min or 1 hr after treatment. Neverthe- ments, for instance, the lowest dose of IL-1␤ (0.025 ␮g) less, given that the effects of IL-1␤ on HPA activation generally did not influence consumption, although a may be augmented by IL-6 in a time-dependent fashion relatively modest but statistically significant reduction (Wang and Dunn 1999), the possibility cannot be ex- of consumption was apparent in the experiment assess- cluded that a synergism would have appeared had ing the combined effects of IL-1␤ and IL-6. other doses or time frames been assessed. Commensurate with earlier reports, both IL-1␤ and It is important to underscore that the present investi- TNF-␣ increased plasma corticosterone concentrations, gation involved administration recombinant human cy- whereas IL-6 did not elicit such an outcome (Connor et tokines, rather than murine cytokines. Human and mu- al. 1998; Zalcman et al. 1994). At the 0.05 ␮g dose of IL- rine TNF-␣ (hTNF-␣ and mTNF-␣) have functionally 1␤, the level of plasma corticosterone was elevated (An- different effects in mice, as the latter excite both the p55 isman and Merali 1999; Lacosta et al. 1998), whereas at and the p75 receptors, whereas the former excites only the 0.025 ␮g dose, the observed effects, as in the case of the p55 receptor. It has been suggested that the differen- the consumption studies, were more variable across ex- tial effects on these receptors may account for the far periments. Likewise, while the 1.0 ␮g dose of TNF-␣ greater lethality elicited by the murine form of TNF-␣ was uniformly without effect on plasma corticosterone (Brouckaert et al. 1989a). It is significant that the effects levels, at this low dose significant elevations of the hor- of hTNF-␣ can be augmented by the co-administration mone have been observed (Anisman et al. 1998; Hayley of a “sensitizing” agent, so that the effects of this form et al. 1999). It is important to recognize, however, that of the cytokine will be comparable to that elicited by the in the experiments assessing the synergistic effects of mTNF-␣. Among other agents, IL-1 was particularly ef- the cytokines, the doses used were near threshold for fective in enhancing the effects of hTNF-␣ (Brouckaert HPA activation, and hence the corticosterone response et al. 1989b; Everaerdt et al. 1989). Thus, the possibility may have been subject to greater variability across ex- ought to be considered that the synergistic effects of periments. hIL-1␤ and hTNF-␣ stem from direct or indirect activa- In assessing the potential synergistic effects of the cy- tion of the p75 receptor, or actions that parallel the con- tokines, only a single dose of each cytokine was used sequences of the conjoint activation of the p55 and p75 based on the results of the initial studies that evaluated receptors. If this were the case, then it would be ex- the dose-dependent effects of the treatments. However, pected that synergistic effects would not be evident us- as indicated by Berenbaum (1989), analysis of synergy ing the murine form of TNF-␣, as both receptors will al- may require multiple doses in order to assess shifts of ready have been activated. the dose response curve. This is particularly the case with respect to cytokines, as their neurochemical ac- Central Neurochemical Variations tions are not necessarily linear and may be biphasic (Alonso et al. 1993; Lapchak 1992). Coupled with the The neurochemical variations induced by IL-1␤, IL-6, fact that the cytokine doses used in the present investi- and TNF-␣ were particularly intriguing in light of the gation were at near-threshold levels, the results of the suggestion that systemic stressors, such as cytokines, present investigation need to be considered cautiously, may influence HPA activity primarily via non-limbic and the conclusions drawn are provisional. With re- mechanisms (Herman and Cullinan 1997). There have spect to consumption of a favored food substance, it ap- been several reports indicating that cytokine adminis- peared that IL-6 did not enhance the effects of either IL- tration influences hypothalamic monoamine activity 1␤ or TNF-␣. In contrast, the co-administration of be- (Shintani et al. 1993), most notably within the PVN haviorally sub-effective doses of IL-1␤ and TNF-␣ ap- (Dunn 1988, 1992b). As well, IL-1␤ and TNF-␣ also in- peared to synergistically influence consumption, as creased amine activity at some extrahypothalamic sites previously observed with respect to feeding patterns (Ignatowski and Spengler 1994; Linthorst et al. 1995; (Plata-Salaman et al. 1996; Van der Meer et al. 1995; Merali et al. 1997; Song et al. 1999), but few studies as- Yang et al. 1994). sessed the actions within mesolimbic regions, and none Paralleling the behavioral effects, co-administration evaluated potential synergistic effects. It is clear from of IL-1␤ and TNF-␣ appeared to synergistically increase the present investigation, summarized in Table 5, that plasma corticosterone concentrations, whereas IL-6 did IL-1␤ and TNF-␣ influenced monoamine turnover in not augment the action of either these cytokines. While both limbic and non-limbic sites. Relatively low doses IL-1␤ and IL-6 synergistically influenced ACTH and of IL-1␤ elevated MHPG, and 5-HIAA within the PVN, corticosterone, depending on the timing of their admin- increased NE turnover at the locus coeruleus and cen- istration (Perlstein et al. 1991; Zhou et al. 1996), such an tral amygdala, 5-HT turnover within the central outcome was not apparent in the present investigation. amygdala and prefrontal cortex, as well as DA utiliza- Moreover, in independent experiments there was no tion within the prefrontal cortex. In contrast, IL-1␤ had evidence of a synergism irrespective of whether blood scant effects on hippocampal monoaminergic activity, NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 577

Table 5. Effect of Cytokine Administration on the Levels of present investigation remains to be elucidated, and it is Biogenic Amines in Various Brain Regionsa uncertain whether the amine changes were responsible for the neuroendocrine variations. Although the locus IL-1␤ coeruleus sends only sparse projections to the PVN NE/MHPG 5-HT/5-HIAA DA/DOPAC (Cunningham and Sawchenko 1988), it may be involved in the mediation of HPA activity (Gann et al. PVN 0/↑↑/↑ 0/0 LC 0/↑ ——1977). A wide range of stressors influence locus coer- CEA 0/↑ 0/↑ 0/0 uleus activity (Abercrombie and Jacobs 1987; Weiss PFC 0/0 ↑/↑↑/↑ 1991), and it is conceivable that the HPA alterations in- HIP 0/0 0/0 — duced by the cytokines were related to functional NAC — — 0/0 changes at this site. As well, NE within the PVN may TNF-␣ contribute to HPA alterations elicited by cytokine challenge (Dunn 1988, 1992b; Whitnall 1993). Furthermore, NE/MHPG 5-HT/5-HIAA DA/DOPAC 5-HT has important regulatory actions on HPA func- PVN 0/0 ↓/0 0/0 tioning, just as CRH release may influence forebrain 5- ↓ LC /0 — — HT activity (Dinan 1996; Feldman et al. 1987; Korte et CEA 0/0 0/0 0/0 al. 1991). It is tempting to speculate that the IL-1␤-pro- PFC 0/0 ↑/0 0/0 HIP ↓/0 0/0 — voked increase of NE and 5-HT turnover within the NAC — — 0/↑ PVN, as well as the increased turnover of 5-HT elicited by TNF-␣, may be related to the HPA activation in- Arrows indicate significant increases or decreases from baseline levels. duced by these cytokines. 0 indicates no change from baseline levels. aIL-6 produced a significant increase of both 5-HT and DA levels in the Although IL-1␤ and TNF-␣ had clear neuroendo- central nucleus of the amygdala. No other changes were evident in any crine and central neurochemical effects, and appeared brain region analyzed. to act synergistically with respect to food consumption and plasma corticosterone levels, there was no evidence although in vivo studies have indicated that IL-1␤ at of a synergism with respect to central monoamine activ- higher doses enhanced 5-HT activity (Linthorst et al. ity. Clearly, the joint actions of these cytokines on cer- 1995; Merali et al. 1997). tain aspects of endocrine functioning are not necessarily The effects of TNF-␣ were less widespread than those translated into central monoamine changes. It may be of IL-1␤, and tended to be more variable between experi- that the cytokine doses in the present investigation ments. Levels of 5-HT were affected within the PVN and were not optimal to provoke synergistic central amine prefrontal cortex, whereas NE levels were reduced changes, particularly in view of the limited IL-1␤ con- within the locus coeruleus and the dorsal hippocampus. centrations that likely reach the brain (Hopkins and Yet, the turnover of NE within the locus coeruleus varied Rothwell 1995). Furthermore, as indicated earlier, anal- between experiments, being marked in the study exam- ysis of synergism might have been more profitably as- ining the conjoint effects of TNF-␣ and IL-6, but absent in sessed using a variety of cytokine doses. Indeed, in the the dose response study and in the analysis of the com- present investigation the greatest amine changes were bined effects of IL-1 and TNF-␣. The source for these be- not necessarily elicited by the highest cytokine doses. tween-experiment differences is not evident, although it Furthermore, within the amygdala the elevation of should be recognized that the doses of the human recom- MHPG associated with IL-1␤ were attenuated by the binant TNF-␣ employed in the present investigation co-administration of IL-6. were relatively modest and hence may have yielded lim- Although the source for such an antagonistic effect ited effects. Finally, in contrast to IL-1␤, it was observed cannot be determined from the present investigation, that IL-6 had minimal effects, although it did increase 5- the diminished response associated with the combina- HT and DA levels within the central amygdala. Com- tion of treatments may have been related to the nature mensurate with the findings of Terao et al. (1993), neither of the dose response curves associated with the individ- IL-6 nor TNF-␣ significantly altered hypothalamic NE ual cytokines. Alternatively, it is possible that synergis- activity. However, the neurochemical effects of TNF-␣ tic effects within the brain were limited by strict regula- may depend upon the particular hypothalamic nuclei in- tory processes. For instance, treatment with a second vestigated. In fact, TNF-␣ was shown to dose-depen- cytokine may have promoted negative feedback effects, dently inhibit electrically stimulated NE release from actions on autoreceptors, or synthesis of endogenous axon terminals of isolated rat median eminence and may receptor antagonists (e.g., IL-1ra) which would have act in a modulatory capacity with respect to CRH release acted against the actions of IL-1␤. Finally, in the present from this site (Elenkov et al. 1992). study monoamine variations were assessed in post- The relationship between the central neurochemical mortem tissues, and it is certainly likely that in vivo de- and the corticosterone variations observed in the terminations of amine release would have provided a 578 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

better index of the dynamic effects associated with cy- Endocrine and cytokine correlates of major depression tokine treatment. In fact, synergistic effects between cy- and dysthymia with typical or atypical features. Mol tokines and stressors have been detected in vivo (Merali Psychiatry 4:182–188 et al. 1997). Berenbaum MC (1989): What is synergy? Pharmacol Rev The finding that central monoamine activity was in- 41:93–141 fluenced by cytokines is consistent with the view that Blalock JE (1994): The syntax of immune-neuroendocrine the immune system and the brain interact, and lends communication. Immunol Today 15:504–511 support to the potential psychological effects associated Bluthe RM, Dantzer R, Kelley KW (1992): Effects of interleu- with such treatments (Maier and Watkins 1998). In this kin-1 receptor antagonist on the behavioral effects of lipopolysaccharide in the rat. Brain Res 575:318–320 respect, it was shown that behavioral changes (e.g., lo- comotor activity) may correspond with in vivo amine al- Bluthe RM, Pawlowski M, Suarez S, Parnet P, Pittman Q, terations elicited by cytokines and by endotoxin admin- Kelley KW, Dantzer R (1994): Synergy between tumor necrosis factor-␣ and interleukin-1 in the induction of istration (Linthorst et al. 1995; Linthorst and Reul 1998). sickness behavior in mice. Psychoneuroendocrinology In addition, such treatments may impact on anxiogenic 19:197–207 and anhedonic processes (Borowski et al. 1998; Connor Borowski T, Kokkinidis L, Merali Z, Anisman H (1998): et al. 1998), both of which are related to depressive ill- Lipopolysaccharide, central in vivo amine alterations, ness. Indeed, the finding that IL-1␤ and TNF-␣ influ- and anhedonia. NeuroReport 9:3797–3802 enced consumption of a palatable substance, increased Brouckaert PG, Everaerdt B, Libert C, Takahashi N, Fiers W HPA activity, and altered monoamine turnover in brain (1989a): Species specificity and involvement of other regions thought to be affected by stressors, is commen- cytokines in endotoxin shockaction of recombinant surate with the proposition that these cytokines may be tumor necrosis factor in mice. Agents Actions 26:196– 198 related to depressive illness (Maes 1995). It is significant, as well, that in human trials the use of some cyto- Brouckaert PG, Libert C, Everaerdt B, Takahashi N, Fiers W (1989b): A role for interleukin-1 in the in vivo actions of kines, such as IL-2, have been found to provoke marked tumor necrosis factor. Lymphok Res 8:269–274 psychiatric and cognitive disturbances (Capuron et al. Capuron L, Ravaud A, Radat F, Dantzer R, Goodall G (1998): 1998; Caraceni et al. 1992; Denicoff et al. 1987) suffi- Affects of interleukin-2 and alpha-interferon cytokine ciently severe to require discontinuation of treatment immunotherapy on the mood and cognitive perfor- (Meyers and Valentine 1995). Clearly, a much better un- mance of cancer patients. Neuroimmunomodulation 5:9 derstanding of cytokine effects on central neurochemi- Caraceni A, Martini C, Belli F, Mascheroni L, Rivoltini L, Ari- cal functioning and on cognitive processes is necessary enti F, Cascinelli N (1992): Neuropsychological and neu- in order to optimize the efficacy and tolerability of im- rophysiological assessment of the central effects of munotherapy. interleukin-2 administration. Eur J Cancer 29A:1266–1269 Chiueh CC, Zukowska-Grojec Z, Kirk KL, Kopin JJ (1983): 6- fluorocatechaolamine as a false adrenergic neurotransmitter. J Pharmac Exp Ther 225:529–533 ACKNOWLEDGMENTS Connor TJ, Song C, Leonard BE, Merali Z, Anisman H (1998): An assessment of the effects of central interleu- Supported by a grant in aid of research from the Medical Re- kin-1␤, -2, -6 and tumor necrosis factor-␣ administra- search Council of Canada. H.A. is an Ontario Mental Health tion on some behavioral, neurochemical, endocrine and Senior Research Fellow. The assistance of Jerzy Kulczycki is immune parameters in the rat. Neuroscience 84:923–933 very much appreciated. Cunningham ET Jr, Sawchenko PE (1988): Anatomical specificity of noradrenergic inputs to the paraventricular and supraoptic nuclei of the rat hypothalamus. J Comp Neu- REFERENCES rol 274:60–76 Dantzer R, Bluthe RM, Auber A, Goodall G, Bret-Dibat JL, Abercrombie ED, Jacobs BL (1987): Single-unit response of Kent S, Goujon E, Laye S, Parnet P, Kelley KW (1996): noradrenergic neurons in the locus coeruleus of freely Cytokine actions on behavior. In Rothwell NJ (ed), moving cats. I. Acutely presented stressful and non- Cytokines and the Nervous System. London, Landes, stressful stimuli. J Neurosci 9:2837–2843 pp 117–144 Alonso R, Chaudieu I, Diorio J, Krishnamurthy A, Quirion Denicoff KD, Rubinow DR, Papa MZ, Simpson L, Seipp LA, R, Boksa P (1993): Interleukin-2 modulates evoked Lotze MT, Chang AE, Rosenstein D, Rosenberg SA release of [3H]dopamine in rat cultured mesencephalic (1987): The neuropsychiatric effects of treatment with cells. J Neurochem 61:1284–1290 interleukin-2 and lymphokine-activated killer cells. Ann Int Med 107:293–300 Anisman H, Merali Z (1999): Cytokines and stress in relation to anxiety and anhedonia. In Dantzer R, Wollman EE, Dinan T (1996): Serotonin and the regulation of hypothalamic- Yirmiya R (eds), Cytokines, Stress and Depression. New pituitary-adrenal axis function. Life Sci 58:1683–1694 York, Plenum Press Dunn AJ (1988): Systemic interleukin-1 administration stim- Anisman H, Ravindran AV, Griffiths J, Merali Z (1998): ulates hypothalamic norepinephrine metabolism paral- NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6 Synergistic Effects of Cytokines 579

leling the increased plasma corticosterone. Life Sci Kent S, Bret-Dibat JL, Kelley KW, Dantzer R (1996): Mecha- 43:429–435 nisms of sickness induced decreases in food-motivated Dunn AJ (1990): Interleukin-1 as a stimulator of hormone behavior. Neurosci Biobehav Rev 20:171–175 secretion. Prog Neurol Endocrinol Immunol 3:26–34 Korte SM, Van Dunn S, Bouws GA, Koolhaas JM, Bohus B Dunn AJ (1992a): The role of interleukin-1 and tumor necro- (1991): Involvement of hypothalamic serotonin in acti- sis factor alpha in the neurochemical and neuroendo- vation of the sympathoadrenomedullary system hypo- crine responses to endotoxin. Brain Res Bull 6:807–812 thalamo-pituitary-adrenocortical axis in male wistar rats. Eur J Pharmacol 197:225–228 Dunn AJ (1992b): Endotoxin-induced activation of cerebral catecholamine and serotonin metabolism: Comparison Kuemmerle JF (1998): Synergistic regulation of NOS II expres- with interleukin-1. J Pharmac Exp Ther 261:964–969 sion by IL-1 beta and TNF-alpha in cultured rat colonic smooth muscle cells. Am J Physiol 274:G178–G185 Ebisui O, Fukata J, Murakami N, Kobayashi H, Segawa H, Muro S, Hanaoka I, Naito Y, Masui Y, Ohmoto Y, Imura Lacosta S, Merali Z, Anisman H (1999): Behavioral and neu- H, Nakao K (1994): Effect of IL-1 receptor antagonist and rochemical consequences of lipopolysaccharide in mice: antiserum to TNF-␣ on LPS-induced plasma ACTH and Anxiogenic-like effects. Brain Res 818:291–303 corticosterone rise in rats. Am J Physiol 266:E986–E992 Lacosta S, Merali Z, Anisman H (1998): Influence of interleu- Elenkov IJ, Kovacs K, Duda E, Stark E, Vizi ES (1992): Pre- kin-1beta on exploratory behaviors, plasma ACTH, cor- synaptic inhibitory effect of TNF-alpha on the release of ticosterone, and central biogenic amines in mice. noradrenaline in isolated median eminence. J Neuroim- Psychopharmacology 137:351–361 munol 41:117–120 Lapchak PA (1992): A role for IL-2 in the regulation of stri- atal dopaminergic function. NeuroReport 3:165–168 Everaerdt B, Brouckaert PG, Shaw A, Fiers W (1989): Four different interleukin-1 species sensitize to the lethal Linthorst AC, Reul JM (1998): Brain neurotransmission dur- action of tumor necrosis factor. Biochem Biophys Res ing peripheral inflammation. Ann NY Acad Sci Commun 163:378–385 840:139–152 Feldman S, Conforti N, Melamed E (1987): Paraventricular Linthorst ACE, Flachskamm C, Muller-Preuss P, Holsboer F, nucleus serotonin mediates neurally stimulated adreno- Reul JMHM (1995): Effect of bacterial endotoxin and cortical secretion. Brain Res Bull 18:165–168 interleukin-1␤ on hippocampal serotonergic neurotransmission, behavioral activity, and free corticoster- Franklin KBJ, Paxinos G (1997): The Mouse Brain in Stereo- one levels: An in vivo microdialysis study. J Neurosci taxic Coordinates. San Diego, Plenum 15:2920–2934 Gann DS, Ward DG, Baertschi AJ, Carlson DE, Maran JW Long NC, Otterness I, Kunkel SL, Vander AJ, Kluger MJ (1977): Neural control of ACTH release in response to (1990): Roles of inteleukin-1␤ and tumor necrosis factor hemorrage. Ann NY Acad Sci 297:477–497 in lipopolysaccharide fever in rats. Am J Physiol Gayle D, Ilyin SE, Miele ME, Plata-Salaman C (1998): Modu- 259:R724–R728 lation of TNF-alpha mRNA production in rat C6 glioma Lu G, Beuerman RW, Zhao S, Sun G, Nguyen DH, Ma S, cells by TNF-alpha, IL-1beta, IL-6, and IFN-alpha: In Kline, DG (1997): Tumor necrosis factor-alpha and inter- vitro analysis of cytokine-cytokine interactions. Brain leukin-1 induce activation of MAP kinase and SAP Res Bull 47:231–235 kinase in human nueroma fibroblasts. Neurochem Intl Hayley S, Brebner K, Lacosta S, Merali Z, Anisman H (1999): 30:401–410 Sensitization to the effects of tumor necrosis factor-␣; Maes M (1995): Evidence for an immune response in major Neuroendocrine, central monoamine and behavioral depression: A review and hypothesis. Prog Neuro-Psy- variations. J Neurosci 19:5654–5665 chopharmacol Biol Psychiatry 19:11–38 Herman JP, Cullinan WE (1997): Neurocircuitry of stress: Maes M, Meltzer HY, Bosmans E, Bergmans R, Vandoolae- Central control of the hypothalamo-pituitary-adreno- ghe E, Ranjan R, Desnyder R (1995): Increased plasma cortical axis. Trends Neurosci 20:78–84 concentrations of interleukin-6, soluble interleukin-6, Hopkins SJ, Rothwell NJ (1995): Cytokines and the nervous soluble interleukin-2 and transferrin receptor in major system. I. Expression and recognition. Trends Neurosci depression. J Affect Disord 34:301–309 18:83–88 Maier SF, Watkins LR (1998): Cytokines for psychologists: Ignatowski TA, Spengler RN (1994): Tumor necrosis factor- Implications of bidirectional immune-to-brain commu- ␣: Presynaptic sensitivity is modified after antidepres- nication for understanding behavior, mood, and cogni- sant drug administration. Brain Res 665:293–299 tion. Psychol Rev 105:83–107 Jeohn GH, Kong LY, Wilson B, Hudson P, Hong JS (1998): Masana MI, Heyes MP, Mefford IN (1990): Indomethacin Syergistic neurotoxic effects of combined treatments prevents increased catecholamine turnover in rat brain with cytokines in murine primary mixed neuron/glia following systemic endotoxin challenge. Prog Neuro- cultures. J Neuroimmunol 85:1–10 Psychopharmacol Biol Psychiatry 14:609–621 Johnson RW, Propes MJ, Shavit Y (1996): Corticosterone Matta SG, Weatherbee J, Sharp BM (1992): A central mecha- modulates behavioral and metabolic effects of nism is involved in the secretion of ACTH in response lipopolysaccharide. Am J Physiol 270:R192–R198 to IL-6 in rats: Comparison to and interaction with IL- ␤ Kakucksa I, Qi Y, Clark BD, Lechan RM (1993): Endotoxin- 1 . Neuroendocrinology 56:516–525 induced corticotropin-releasing hormone gene expres- Mefford IN, Masters CF, Heyes MP, Eskay RL (1991): Cytok- sion in the paraventricular nucleus is mediated by inter- ine-induced activation of the neuroendocrine stress axis leukin-1. Endocrinology 133:815–821 persists in endotoxin-tolerant mice. Brain Res 557:327–330 580 K. Brebner et al. NEUROPSYCHOPHARMACOLOGY 2000–VOL. 22, NO. 6

Merali Z, Lacosta S, Anisman H (1997): Effects of interleu- Asio S, Kato R, Asai M (1995): Involvement of interleukin-1beta and mild stress on alterations of norepineph- kin-1 immobilization stress-induced increase in adreno- rine, dopamine and serotonin neurotransmission: A corticotropic hormone and in the release of hypothalamic regional microdialysis study. Brain Res 761:225–235 monoamines in the rat. J Neurosci 15:1961–1970 Meyers CA, Valentine AD (1995): Neurological and psychi- Song C, Merali Z, Anisman H (1999): Variations of nucleus atric adverse effects of immunological therapy. CNS accumbens dopamine and serotonin following systemic Drugs 3:56–68 interleukin-1, interleukin-2 or interleukin-6 treatment. Neuroscience 88:823–836 Mino T, Sugiyama E, Taki H, Kuroda A, Yamashita N, Maruyama M, Kobayashi M (1998): Interleukin-1alpha Sonti G, Ilyin SE, Plata-Salaman C (1996): Anorexia induced and tumor necrosis factor alpha syergistically stimulate by cytokine interactions at pathophysiological concen- prostaglandin E2-dependent production of interleukin- trations. Am J Physiol 270:R1394–R1402 11 in rheumatoid synovial fibroblasts. Arthritis Rheum Terao A, Oikawa M, Saito M (1993): Cytokine-induced 41:2004–2013 change in hypothalamic norepinephrine turnover: Muller N, Ackenheil M (1998): Psychoneuroimmunology Involvement of corticotropin-releasing hormone and and the cytokine action in the CNS: Implications for prostaglandins. Brain Res 622:257–261 psychiatric disorders. Prog Neuropsychopharmacol Biol Tilders JJH, DeRijk RH, Van Dam AM, Vincent VAM, Psychiatry 22:1–33 Schotanus K, Persoons JHA (1993): Activation of the O’Reilly B, Vander AJ, Kluger MJ (1987): Effects of chronic hypothalamus-pituitary-adrenal axis by bacterial endot- infusion of lipopolysaccharide on food intake and body oxins: Routes and intermediate signals. Psychoneuroen- temperature of the rat. Physiol Behav 42:287–291 docrinology 19:209–232 Palazzolo DL, Quadri SK (1992): Interleukin-1 inhibits sero- Van der Meer MJM, Sweep CGJF, Pesman GJ, Borm GF, Her- ␤ ␣ tonin release from the hypothalamus in vitro. Life Sci mus RMM (1995): Synergism between IL-1 and TNF- 51:1797–1802 on the activity of the pituitary adrenal axis and on food intake of rats. Am J Physiol 268:E551–E557 Perlstein RS, Moughey EH, Jackson WE, Neta R (1991): Inter- leukin-1 and interleukin-6 act synergistically to stimu- Vogel SN, Henricson BE, Neta R (1991): Roles of interleukin late the release of adrenocorticotropic hormone in vivo. and tumor necrosis factor in lipopolysaccharide-induced Lymphok Cytok Res 10:141–146 hypoglycemia. Brain Behav Immun 59:2494–2498 Plata-Salaman CR (1988): Food intake suppression by imu- Wang J, Dunn AJ (1999): The role of interleukin-6 in the acti- nomodulators. Neurosci Res Comm 3:159–165 vation of the hypothalamo-pituitary-adrenocortical axis and brain indoleamines by endotoxin and interleukin-1 Plata-Salaman CR, Oomura Y, Kai Y (1988): Tumor necrosis beta. Brain Res 815:237–248 factor and interleukin-1␤: Suppression of food intake by direct action in the central nervous system. Brain Res Weiss JM (1991): Stress induced depression: Critical neuro- 448:106–114 chemical and electrophysiological changes. In Madden J (ed), Neurobiology of Learning, Emotion and Affect. Plata-Salaman CR, Sonti G, Borkoski JP, Wilson CD, Ffrench- New York, Raven Press Mullen JMH (1996): Anorexia induced by chronic central administration of cytokines at estimated pathophys- Whitnall MH (1993): Regulation of the hypothalamic corti- iological concentrations. Physiol Behav 60:867–875 cotropin-releasing hormone neurosecretory system. Prog Neurobiol 40:573–629 Rivier C (1993): Effect of perhiperal and central cytokines on Yang Z, Koseki M, Meguid MM, Gleason JR, Debonis D the hypothalamic-pituitary-adrenal axis of the rat. Ann ␣ ␣ NY Acad Sci 697:97–105 (1994): Synergistic effect of rhTNF- and rhIL-1 in inducing anorexia in rats. Am J Physiol 267:R1056–R1064 Rothwell NJ, Allan S, Toulmond S (1997): The role of interleukin-1 in acute neurodegeneration and stroke: patho- Zalcman S, Green-Johnson JM, Murray L, Nance DM, Dyck physiological and therapeutic implications. J Clin Invest D, Anisman H, Greenberg A (1994): Cytokine-specific 100:2648–2652 central monoamine alterations induced by interleukin (IL)-1, IL-2 and IL-6. Brain Res 643:40–49 Seegal RF, Bosh KO, Bush B (1986): High performance liquid Zanetti G, Heusmann D, Gerain J, Kohler J, Abbet P, Barras chromatography of biogenic amines and metabolites in C, Lucas R, Glauser M, Baumgartner J (1992): Cytokine the brain, cerebrospinal fluid, urine and plasma. J Chro- production after intravenous or peritoneal gram nega- matogr 377:131–144 tive bacterial challenge in mice. J Immunol 6:1890–1897 Shintani F, Kanba S, Nakaki T, Nibuya M, Kinoshita N, Zhou D, Shanks N, Riechman SE, Liang R, Kusnecov AW, Suzuki E, Yagi G, Kato R, Asai M (1993): Interleukin-1␤ Rabin BS (1996): Interleukin 6 modulates interleukin-1 augments release of norepinephrine, dopamine and and stress-induced activation of the hypothalamic-pitu- serotonin in the rat anterior hypothalamus. J Neurosci itary-adrenal axis in male rats. Neuroendocrinology 13:3574–3581 63:227–236 Shintani F, Nakaki T, Kanba S, Sato K, Yagi G, Shiozawa M,